This study investigates the effects of adapting elementary science instruction to students' entering achievement levels on three dependent variables: (1) performance on tests of science content knowledge; (2) mastery of lesson objectives; and (3) attitude toward science. Subjects were 235 sixth graders from four elementary schools in a small, mid-western industrial city. Two experimental and two control schools provided instruction on six lessons from the "Science: A Process Approach" (SAPA) curriculum over a ten week period. The experimental group was divided into three relatively homogenous groups based on pretest scores (high, middle or low). An attempt was made to secure mastery by as many students as possible in the experimental groups of the same science objectives. Instruction in control schools was non-differentiated and students were taught in heterogeneous classes. A 2x3 (treatment by group) factorial design was used. Two-way analyses of variance were performed on the data from a 90-item SAPA-content posttest. Students in the high and middle experimental groups attained higher scores on the posttest. Chi-square tests showed a greater number of students mastered the science objectives (80%) in the high and middle experimental groups. (Author/MR)